Information processing systems, information processing methods, information processing programs, and AI agents

Abstract

This invention provides an information processing system, information processing method, information processing program, and AI agent that can predict the sales status of products and the store's layout in drugstores and other retail stores, and execute product replenishment and sales strategies at the appropriate time. [Solution] The system comprises an AI agent and a storage means capable of storing purchase history information, which is information about the history of products purchased by the customer. 、 The AI ​​agent comprises: an image acquisition means capable of acquiring store image data capturing the sales floor conditions of a store; a learning means for learning at least the relationship between the store image data and the purchase history information; and a prediction means capable of predicting the sales floor conditions of a certain store and the sales performance of a certain product in a certain store based on the relationship between the store image data and the purchase history information in that store. Furthermore, the sales floor image data includes image data that allows recognition of the inventory status of products displayed in the store's sales floor, the learning means learns at least the relationship between the inventory status information of a certain product obtained from the sales floor image data and the purchase quantity of that product obtained from the purchase history information, and the prediction means identifies the current inventory status of a certain product by referring to the sales floor image data and predicts the sales status of that product using the purchase quantity of that product associated with the inventory status information of that product.

Description

Technical Field

[0001] The present invention relates to an information processing system, an information processing method, an information processing program, and an AI agent using an AI agent.

Background Art

[0002] Conventionally, there has been proposed a system including means for reading weather data by generative AI and presenting means for predicting product sales and planning product deliveries one week later, means for adjusting the purchases and the number of salespersons of foodstuffs, etc. at each store according to the weather a few days later, and means for formulating event plans for retail stores and hotels for each season based on the weather (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, although the conventional system enables stores such as restaurants and hotels to make efficient purchases and personnel adjustments according to the weather, it has been difficult to apply it to fields such as drugstores.

[0005] In view of such circumstances, the present invention provides an information processing system, an information processing method, an information processing program, and an AI agent that can predict the sales status of products and the storefront situation of stores such as drugstores, and execute replenishment of products displayed in the storefront of the store and product sales measures at appropriate timings.

Means for Solving the Problems

[0006] The information processing system according to the present invention comprises an AI agent and a storage means capable of storing purchase history information, which is information about the history of products purchased by a customer, wherein the AI ​​agent comprises an image acquisition means capable of acquiring store image data capturing the sales floor conditions of a store, a learning means for learning at least the relationship between the store image data and the purchase history information, and a prediction means capable of predicting the sales floor conditions of a certain store and the sales performance of a certain product in a certain store from one of the other, based on the relationship between the store image data and the purchase history information in a certain store. The sales floor image data includes image data that allows recognition of the inventory status of products displayed in the store's sales floor, the learning means at least learns the relationship between the inventory status information of a certain product obtained from the sales floor image data and the purchase quantity of that product obtained from the purchase history information, and the prediction means identifies the current inventory status of a certain product by referring to the sales floor image data and predicts the sales status of that product using the purchase quantity of that product associated with the inventory status information of that product. This is an information processing system characterized by the following:

[0007] The information processing method according to the present invention is an information processing method performed using an AI agent which is a computer and a storage means capable of storing purchase history information which is information about the history of products purchased by a customer, wherein the AI ​​agent performs at least an image acquisition step capable of acquiring store image data which captures the sales floor situation of a store; a learning step which learns at least the relationship between the store image data and the purchase history information; and a prediction step which, based on the relationship between the store image data and the purchase history information at a certain store, can predict the sales floor situation of a certain store and the sales situation of a certain product at a certain store from one of the other. The store image data includes image data that allows recognition of the inventory status of products displayed in the store; the learning step at least learns the relationship between the inventory status information of a certain product obtained from the store image data and the purchase quantity of the certain product obtained from the purchase history information; the prediction step identifies the current inventory status of a certain product by referring to the store image data and predicts the sales status of the certain product using the purchase quantity of the certain product associated with the inventory status information. This is an information processing method characterized by the following features.

[0008] The information processing program according to the present invention is a program for an information processing system comprising an AI agent and a storage means capable of storing purchase history information, which is information about the history of products purchased by a customer, wherein the AI ​​agent, which is a computer, functions as an image acquisition means capable of acquiring store image data capturing the sales floor situation of a store, a learning means for learning at least the relationship between the store image data and the purchase history information, and a prediction means capable of predicting one of the sales floor situation of a certain store and the sales situation of a certain product in a certain store based on the relationship between the store image data and the purchase history information in a certain store. The sales floor image data includes image data that allows recognition of the inventory status of products displayed in the store's sales area, the learning means at least learns the relationship between information on the inventory status of a certain product obtained from the sales floor image data and the purchase quantity of that product obtained from the purchase history information, and the prediction means identifies the current inventory status of a certain product by referring to the sales floor image data and predicts the sales status of that product using the purchase quantity of that product associated with the information on the inventory status of that product. This is an information processing program characterized by the following:

[0009] The AI ​​agent according to the present invention is an AI agent that acts as a substitute for a human and has the ability to learn and make decisions on its own, wherein the AI ​​agent, which is a computer, functions as an image acquisition means capable of acquiring store image data capturing the sales floor situation of a store, a learning means that learns the relationship between at least the store image data and purchase history information which is information about the history of products purchased by a customer, and a prediction means capable of predicting the sales floor situation of a certain store and the sales situation of a certain product in a certain store from one of the other based on the relationship between the store image data and the purchase history information in a certain store. The sales floor image data includes image data that allows recognition of the inventory status of products displayed in the store's sales floor, the learning means at least learns the relationship between the inventory status information of a certain product obtained from the sales floor image data and the purchase quantity of that product obtained from the purchase history information, the prediction means identifies the current inventory status of a certain product by referring to the sales floor image data, and predicts the sales status of that product using the purchase quantity of that product associated with the inventory status information of that product. The learning means is an AI agent characterized by being configured to learn the relationship between the sales floor image data and the purchase history information, which are repeatedly provided as learning data. [Effects of the Invention]

[0010] The information processing system, information processing method, information processing program, and AI agent according to the present invention can predict the sales status of products and the sales floor conditions of stores such as drugstores, and can achieve the excellent effect of replenishing products displayed on the sales floor and implementing sales strategies for products at the appropriate time. [Brief explanation of the drawing]

[0011] [Figure 1] This is a schematic diagram showing an overview of the information processing system 10 according to this embodiment 1. [Figure 2] This is a system configuration diagram showing an example of the configuration of the information processing system 10 according to this embodiment 1. [Figure 3] This diagram shows an example of a store's sales floor situation and a prediction of product sales. [Figure 4] This is a schematic diagram showing an overview of the information processing system 10 according to this second embodiment. [Figure 5] This is a system configuration diagram showing an example of the configuration of the information processing system 10 according to this second embodiment. [Figure 6] It is a diagram showing an example of prediction data in a certain store and the prediction of the store's sales floor situation.

Mode for Carrying Out the Invention

[0012] <<Embodiment 1>> Hereinafter, the information processing system 10 according to Embodiment 1 of the present invention will be described with reference to FIGS. 1 to 3.

[0013] <Overview of the Information Processing System> First, the overview of the information processing system 10 according to Embodiment 1 will be described with reference to FIG. 1. FIG. 1 is a schematic diagram showing the overview of the information processing system 10 according to Embodiment 1.

[0014] The information processing system 10 according to the present embodiment is an information processing system configured to include an AI agent 12, an AI agent, and a storage means 14 capable of storing purchase history information, which is information on the history of products purchased by customers. The AI agent 12 includes an image acquisition means 12a capable of acquiring sales floor image data obtained by imaging the sales floor situation of a store, a learning means 12b for learning at least the relevance between the sales floor image data and the purchase history information, and a prediction means 12c capable of predicting one of the sales floor situation of a certain store and the sales situation of the products in the certain store from the other based on the relevance between the sales floor image data and the purchase history information in the certain store. It is an information processing system characterized by the above.

[0015] According to the information processing system according to the present embodiment, it is possible to predict the sales situation of products and the sales floor situation in a store such as a drugstore, and execute the replenishment of products displayed on the store's sales floor and the sales measures of products at an appropriate timing.

[0016] Here, the "AI agent" is a system that operates as an agent for humans, has the ability to learn and make judgments on its own, and is a system that integrates various AI technologies for solving complex problems.

[0017] The AI agent executes processing based on an activation trigger given from the outside (e.g., a question (prompt) by the user, a signal input from a sensor), or executes processing based on an activation trigger given from the inside (e.g., activation by an internal scheduler, occurrence of an abnormality).

[0018] For example, when a question (prompt) is given as an activation trigger (event) from the outside, tasks such as searching for an answer to the question or generating an answer are executed, and the answer to the question is output. Also, for example, when a task (prompt) is given as an activation trigger (event) from the outside, tasks such as searching for a solution to the task or generating a solution are executed, and the solution to the task is output.

[0019] Moreover, a "prompt" refers to an instruction or question given by a user to the system, and includes, for example, an AI prompt given to a generative AI, a command prompt that gives an instruction by a command sentence (command), etc.

[0020] Various processes executed by the AI agent can also be executed by "agent-based AI".

[0021] Note that the "activation trigger" is not limited to an activation trigger given based on an event from the outside (e.g., a question (prompt) by the user, a signal input from a sensor), and may be an activation trigger given based on an event from the inside (e.g., activation by an internal scheduler, occurrence of an abnormality).

[0022] Various processes executed by the AI agent can also be executed by "agent-based AI".

[0023] Here, "agent-based AI" refers to a system in which multiple AI agents work together. In agent-based AI, each AI agent takes on a specialized role to solve complex problems that a single AI agent cannot solve, exchanging information with each other to output answers to questions or solutions to problems.

[0024] "Purchase history information" refers to information about the history of products purchased by a specific individual. Examples of "purchase history information" include: (1) customer information (customer ID, gender, age, membership rank, etc.), (2) product information (product ID, product name, category, brand, unit price, etc.), (3) purchase information (purchase date, purchase time, purchase quantity, total amount, payment method, etc.), (4) store information (store ID, store name, sales channel (store / e-commerce), etc.), (5) campaign information (coupon usage, point usage, discount rate, etc.), and (6) purchase frequency (number of purchases, average purchase interval, most recent purchase date, etc.).

[0025] "Sales floor image data" refers to image data that allows for the recognition of the "sales floor conditions" of a store. "Sales floor conditions" include the number, condition, or name of products displayed on shelves in the sales floor, the availability of shelves in the sales floor, the number, condition, or name of products placed near shelves in the sales floor, the layout of shelves in the sales floor, the number of shoppers in the sales floor, and the status of store staff in the sales floor.

[0026] Examples of "store images" include images that allow recognition of increases or decreases in the number of products displayed on shelves in the store, images that allow recognition of the number, condition, or product name of products displayed on shelves in the store, images that allow recognition of the availability of shelves in the store, images that allow recognition of the number, condition, or product name of products placed near shelves in the store, images that allow recognition of the layout of shelves in the store, images that allow recognition of the number of shoppers in the store, and images that allow recognition of the status of store staff in the store.

[0027] The term "relevance" includes both "correlation" and "causation." "Correlation" refers to a relationship where one thing changes simultaneously with the other. For example, the "correlation" between store image data and purchase history information would be that when store image data indicates a decrease in the number of a certain product displayed on store shelves, the purchase quantity of that product in the purchase history information increases. "Causation" refers to a relationship where a change in one thing causes a change in the other. For example, the "causal relationship" between store image data and purchase history information would be that when store image data indicates a decrease in the number of a certain product displayed on store shelves, the purchase quantity of that product in the purchase history information increases.

[0028] Furthermore, the system may include means for acquiring external factors related to a particular store (for example, weather information for the area including the location of the store, the presence or absence of competitors to the store, the selling price of goods at competitors to the store, information on events or school activities in the area where the store is located, disaster information in the area where the store is located, and information on the trade area and population of the area where the store is located), and the prediction means 12c may predict the sales performance of goods at a particular store based on the sales floor conditions of the store and the external factors of the store.

[0029] With this configuration, it becomes possible to predict the sales performance of a particular store by taking into account external factors related to that store, thereby improving the accuracy of the sales performance prediction for that store.

[0030] <Example of system configuration> Next, an example of the configuration of the information processing system 10 according to this second embodiment will be described using Figure 2. Figure 2 is a system configuration diagram showing an example of the configuration of the information processing system 10 according to this second embodiment.

[0031] The information processing system 10 can be configured, for example, to include a system terminal 12 that controls the entire system, a purchasing information terminal 14 that is connected to the system terminal 12 via a network NW so that they can communicate with each other, and an external terminal 16.

[0032] The system terminal 12 is a terminal that controls the entire information processing system 10, and is composed of conventionally known servers, personal computers, etc. In this example, the system terminal 12 is composed of one server, but it may be composed of multiple servers, personal computers, etc. The hardware configuration of the system terminal 12 and the programs that the system terminal 12 executes will be described later.

[0033] The purchase information terminal 14 is a terminal (storage means) for a retailer of goods (for example, a drugstore, retail store, convenience store, etc.) to store purchase history information, and is composed of conventionally known servers, personal computers, etc.

[0034] In this example, the purchase information terminal 14 is configured with a single server, but it may also be configured with multiple servers or personal computers. Furthermore, the terminal (storage means) for storing purchase history information may be an internal storage means (for example, the storage device 26 shown in Figure 2) connected to the system terminal 12 via a local network (for example, LAN or P2P).

[0035] External terminals 16 are terminals used by users of the information processing system 10 (for example, business owners of drugstores, retail stores, convenience stores, etc.), and consist of personal computers, tablets, smartphones, etc. The type of external terminal 16 is not particularly limited, but examples include smartphones, personal computers, tablets, etc. used by individuals.

[0036] The network NW is a line that allows the system terminal 12, the purchasing information terminal 14, and the external terminal 16 to communicate with each other, and is typically composed of a WAN (Wide Area Network), also known as the Internet. The network NW may be wired or wireless, a LAN (Local Area Network), a dedicated line such as a VPN (Virtual Private Network), or a combination of these lines.

[0037] <System Terminal / Hardware Configuration Example> Next, we will describe an example of the hardware configuration of system terminal 12.

[0038] As shown in Figure 2, the system terminal 12 is configured to include, for example, a CPU 21, and a ROM 22, RAM 23, external storage drive 25, storage device 26, input device 27, display device 28, communication unit 29, imaging means 17, etc., all connected to the CPU 21 via a bus.

[0039] The CPU 21 is a control means that controls the entire system terminal 12, and performs processes such as executing application programs and operating systems (OS) stored in ROM 22 and storage devices 26, and storing data and files necessary for program execution in RAM 23 and storage devices 26.

[0040] ROM22 is a storage means for storing basic I / O programs and various data, and is composed of, for example, PROM, flash memory, etc. RAM23 is a storage means for temporarily storing data, and is composed of, for example, SDRAM, DRAM, etc. External storage drive25 is a control means that can read and write data to recording media 24 such as magnetic tape, DVD, etc., and is composed of, for example, magnetic tape storage, DVD drive, etc.

[0041] The storage device 26 is a storage means for storing application programs, the OS, control programs, related programs, various information, etc., and is composed of, for example, a hard disk drive (HDD), a solid-state drive (SDD), etc. The input device 27 is for inputting commands (instructions), etc., to the system terminal 12, and is composed of, for example, a keyboard, a pointing device (mouse, etc.), a touch panel, etc.

[0042] The display device 28 displays commands input by the input device 27, the response output of the system terminal 12 to those commands, and various other displays, and is composed of, for example, a liquid crystal display, plasma display, organic EL, etc. The communication unit 29 is a control means that controls communication with the management information terminal 14, web server 18, external terminal 16, etc. via the network NW, and is composed of, for example, a communication card, etc.

[0043] The imaging means 17 is a means for capturing images of the store's sales floor that allow for recognition of the store's sales floor conditions and transmitting the sales floor image data to the system terminal 12. In this example, it consists of a camera installed facing the display shelves in the store's sales floor.

[0044] <System Terminal / Function> Next, we will explain the functions of the system terminal 12.

[0045] The storage device 26 of the system terminal 12 stores a program (information processing program) that enables the system terminal 12 to function as an image acquisition means 12a, a learning means 12b, and a prediction means 12c.

[0046] <System terminal / Function / Image acquisition method> Next, the image acquisition means 12a will be described.

[0047] The image acquisition means 12a is a means capable of acquiring sales floor image data captured from images of the sales floor conditions of the store, and in this example, it is composed of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0048] As mentioned above, "sales floor image data" refers to image data that allows for the recognition of the store's "sales floor conditions." "Sales floor conditions" include the number, condition, and name of products displayed on shelves, the availability of shelves, the number, condition, and name of products placed near shelves, the layout of shelves, the number of shoppers in the sales floor, and the status of store staff in the sales floor.

[0049] Examples of "store images" include images that can recognize the increase or decrease in a certain product on a display shelf in a store, images that can recognize the number, condition, or product name of a product on a display shelf in a store, images that can recognize the availability of a display shelf in a store, images that can recognize the number, condition, or product name of a product placed near a display shelf in a store, images that can recognize the layout of a display shelf in a store, images that can recognize the number of shoppers in the store, and images that can recognize the status of store staff in the store. Conventional image recognition techniques can be applied to recognize these store images.

[0050] The image acquisition means 12a acquires sales floor image data from an imaging means 17 installed in a certain store, associates the sales floor image data with the ID of that store, and stores it in the storage device 26.

[0051] <System terminal / Functions / Learning methods> Next, we will explain learning method 12b.

[0052] The learning means 12b is a means for learning at least the relationship between store image data and purchase history information, and in this example, it consists of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0053] The learning means 12b refers to the store image data stored in the storage device 26 by the image acquisition means 12a and the purchase history information stored in the purchase information terminal 14, and acquires store image data associated with a certain store ID and purchase history information associated with a certain store ID, and learns the relationship between the store image data and purchase history information at a certain store.

[0054] In other words, the learning means 12b is repeatedly provided with storefront image data associated with a store ID and purchase history information associated with a store ID as learning data (or training data), thereby learning the relationship between storefront image data and purchase history information at a given store.

[0055] For example, consider a case in Figure 3 where, at the date and time indicated by symbol T1, store image data is obtained that shows that only the beverage inventory on the display shelves is low (the beverage shelves are nearly empty), and purchase history information is obtained that shows that beverage sales are low (for example, the quantity of beverages purchased is small).

[0056] In this case, the learning means 12b recognizes that there is a correlation between the information obtained from the store image data, which states that "only the stock of beverages on the display shelves is low," and the purchase history information, which states that "sales of beverages are low." It then stores this information in the storage device 26, as a correlation between the store image data and the purchase history information at a particular store, and associates it with the ID of that store.

[0057] On the other hand, if, at the date and time indicated by code T1, purchase history information is obtained that indicates high sales of beverages (for example, a large quantity of beverages purchased), the learning means 12b recognizes a correlation between the information obtained from the store image data, which indicates that "only the stock of beverages on the display shelves is low," and the purchase history information, which indicates that "sales of beverages are high." The learning means 12b then stores this information in the storage device 26, as a correlation between store image data and purchase history information at a particular store, and associates it with the ID of that store.

[0058] Furthermore, consider a case where, for example, at the date and time indicated by symbol T2 in Figure 3, store image data is acquired that allows recognition that products are lined up without any gaps on the display shelves (the display shelves are full), and purchase history information is acquired that indicates that the overall sales of the products are low (for example, that the quantity of products purchased is small).

[0059] In this case, the learning means 12b recognizes that there is a correlation between the information obtained from the store image data, which states that "products are lined up without any gaps on the shelves," and the purchase history information, which states that "overall sales of the products are low." It then stores this information in the storage device 26, as a correlation between the store image data and the purchase history information at a particular store, and associates it with the ID of that store.

[0060] On the other hand, if, at the date and time indicated by code T2, purchase history information is obtained that indicates a high overall sales volume for the product (for example, a large quantity of the product was purchased), the learning means 12b recognizes a correlation between the information obtained from the store image data, such as "the shelves are filled with products," and the purchase history information, such as "the overall sales volume for the product is high." The learning means 12b then stores this information in the storage device 26, as a correlation between the store image data and purchase history information at a particular store, and associates it with the ID of that store.

[0061] Furthermore, consider a case where, for example, at the date and time indicated by symbol T3 in Figure 3, store image data is acquired that allows it to be recognized that beverages are evenly distributed on the display shelves, and purchase history information is acquired that indicates that beverage sales are high (for example, a large quantity of beverages purchased).

[0062] In this case, the learning means 12b recognizes that there is a correlation between the information obtained from the store image data, which states that "beverages are evenly distributed on the shelves," and the purchase history information, which states that "beverage sales are high." It then stores this information in the storage device 26, as a correlation between the store image data and the purchase history information at a particular store, and associates it with the ID of that store.

[0063] On the other hand, if, at the date and time indicated by code T3, purchase history information is obtained that indicates low sales of beverages (for example, a small quantity of beverages purchased), the learning means 12b recognizes a correlation between the information obtained from the store image data, such as "beverages are evenly distributed on the shelves," and the purchase history information, such as "low sales of beverages," and stores this information in the storage device 26, as a correlation between store image data and purchase history information at a certain store, associated with the ID of that store.

[0064] <System terminal / Function / Prediction method> Next, the prediction means 12c will be described.

[0065] The prediction means 12c is a means capable of predicting the sales situation of a store and the sales situation of a store based on the relationship between sales floor image data and purchase history information of a store, and in this example, it is composed of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0066] The prediction means 12c refers to the relationship between store floor image data and purchase history information stored in the storage device 26 by the learning means 12b, and predicts the other from either the store floor situation or the sales situation of the products at that store.

[0067] For example, as shown by symbol T1 in Figure 3, consider a case where, in relation to store floor image data and purchase history information, information obtained from the store floor image data, such as "only the stock of beverages on the display shelves is low," and purchase history information, such as "sales of beverages are low," are stored.

[0068] In this case, the prediction means 12c refers to the store image data stored in the storage device 26 by the image acquisition means 12a. For example, if the latest store image data stored includes store image data that shows that only the stock of beverages on the shelves is low (the beverage shelves are nearly empty), then the prediction means 12c can predict the sales situation of the product, such as "sales of beverages will decline," based on the current store situation of "only the stock of beverages on the shelves is low," using the relationship between the store image data and purchase history information at that store.

[0069] As a result, users of the information processing system 10 (for example, managers of drugstores, retail stores, convenience stores, etc.) can take measures such as reducing orders for beverages or decreasing the number of beverages shipped from the warehouse based on predictions of product sales, such as "beverage sales will decline," thereby maintaining an appropriate number of beverages on display in the store.

[0070] On the other hand, the prediction means 12c can refer to the purchase history information stored in the purchase information terminal 14 in real time. For example, if the purchase history information includes purchase information indicating that beverage sales are low (e.g., a small quantity of beverages purchased), the prediction means 12c can predict the sales situation, such as "only the beverage inventory on the shelves is decreasing," based on the relationship between the sales image data and the purchase history information at the store.

[0071] Furthermore, as shown by symbol T2 in Figure 3, consider a case where, in relation to store floor image data and purchase history information, information obtained from store floor image data such as "products are lined up without gaps on the shelves" and purchase history information such as "overall sales of products are low" are stored.

[0072] In this case, the prediction means 12c refers to the store image data stored in the storage device 26 by the image acquisition means 12a. For example, if the latest store image data stored is store image data that shows products are lined up without gaps on the shelves (the shelves are full), then the prediction means 12c can predict the sales situation of the products, such as "overall sales of products will decline," based on the relationship between the store image data and purchase history information at that store, from the current store situation of "products are lined up without gaps on the shelves."

[0073] As a result, users of the information processing system 10 (for example, managers of drugstores, retail stores, convenience stores, etc.) can take measures such as refraining from ordering products or reducing the number of products sent out from the warehouse, based on predictions of overall product sales, that "overall product sales will decline," thereby maintaining an appropriate number of products on display in the store.

[0074] On the other hand, the prediction means 12c can refer to the purchase history information stored in the purchase information terminal 14 in real time. For example, if the purchase history information includes purchase information indicating that overall sales of the product are low (e.g., a small quantity of the product is purchased), the prediction means 12c can predict the sales floor situation, such as "the shelves are filled to capacity (the shelves are full)," from the current sales situation of "overall sales of the product are poor," based on the relationship between the sales floor image data and the purchase history information at the store.

[0075] Furthermore, consider, for example, as shown by symbol T3 in Figure 3, a case where, in relation to store floor image data and purchase history information, information obtained from the store floor image data, such as "beverages are evenly distributed on the shelves," and purchase history information, such as "beverage sales are high," are stored.

[0076] In this case, the prediction means 12c refers to the store image data stored in the storage device 26 by the image acquisition means 12a. For example, if the latest store image data stored is store image data that shows that beverages are evenly distributed on the shelves, the prediction means 12c can predict the sales status of the product, such as "beverage sales will improve," based on the relationship between the store image data and purchase history information at that store, from the current store situation, such as "beverages are evenly distributed on the shelves."

[0077] This allows users of the information processing system 10 (for example, managers of drugstores, retail stores, convenience stores, etc.) to take measures such as ordering more beverages or taking beverages out of the warehouse in advance, based on predictions of product sales, such as "beverage sales will be good," thereby replenishing beverages on the sales floor at the appropriate time.

[0078] On the other hand, the prediction means 12c can refer to the purchase history information stored in the purchase information terminal 14 in real time. For example, if the purchase history information includes purchase information indicating high sales of beverages (e.g., a large quantity of beverages purchased), the prediction means 12c can predict the sales floor situation, such as "beverages are selling well," from the current sales situation, such as "beverages are selling well," based on the relationship between the sales floor image data and the purchase history information at the store, and predict the sales floor situation, such as "beverages are remaining evenly on the shelves."

[0079] This example demonstrates that it is possible to predict the sales trends and store layouts of products in drugstores and other retail stores, and to replenish products on the shelves and implement sales strategies at the appropriate time.

[0080] <Information Processing Systems / Summary> As described above, the information processing system according to this embodiment (for example, the information processing system 10 shown in Figures 1 and 2) is an information processing system configured to include an AI agent (for example, the AI ​​agent shown in Figure 1, the system terminal 12 shown in Figure 2) and a storage means (for example, the purchase information terminal 14 shown in Figures 1 and 2) capable of storing purchase history information, which is information about the history of products purchased by a customer, wherein the AI ​​agent is characterized by comprising: an image acquisition means (for example, the image acquisition means 12a shown in Figures 1 and 2) capable of acquiring sales floor image data that captures the sales floor situation of a store; a learning means (for example, the learning means 12b shown in Figures 1 and 2) for learning at least the relationship between the sales floor image data and the purchase history information; and a prediction means (for example, the prediction means 12c shown in Figures 1 and 2) capable of predicting the sales floor situation of a certain store and the sales situation of a certain product in a certain store from one of them, based on the relationship between the sales floor image data and the purchase history information in a certain store.

[0081] Furthermore, the information processing method according to this embodiment (for example, the method executed by the information processing system 10 shown in Figures 1 and 2) is an information processing method that is executed using an AI agent which is a computer (for example, the AI ​​agent shown in Figure 1, the system terminal 12 shown in Figure 2) and a storage means which is capable of storing purchase history information which is information of the history of products purchased by a customer (for example, a purchase information terminal 14 shown in Figures 1 and 2), and is characterized by executing at least an image acquisition step which is capable of acquiring sales floor image data which captures the sales floor situation of a store (for example, a process executed by the image acquisition means 12a shown in Figures 1 and 2), a learning step which is capable of learning the relationship between at least the sales floor image data and the purchase history information (for example, a process executed by the learning means 12b shown in Figures 1 and 2), and a prediction step which is capable of predicting the sales floor situation of a certain store and the sales situation of a certain product in a certain store from one of them based on the relationship between the sales floor image data and the purchase history information in a certain store (for example, a process executed by the prediction means 12c shown in Figures 1 and 2).

[0082] Furthermore, the information processing program according to this embodiment (for example, the program executed by the information processing system 10 shown in Figures 1 and 2) is a program for an information processing system configured to include an AI agent (for example, the AI ​​agent shown in Figure 1, the system terminal 12 shown in Figure 2) and a storage means (for example, the purchase information terminal 14 shown in Figures 1 and 2) capable of storing purchase history information, which is information about the history of products purchased by a customer. The program is characterized in that it causes the AI ​​agent, which is a computer, to function as an image acquisition means (for example, the image acquisition means 12a shown in Figures 1 and 2) capable of acquiring sales floor image data that captures the sales floor situation of a store, a learning means (for example, the learning means 12b shown in Figures 1 and 2) that learns at least the relationship between the sales floor image data and the purchase history information, and a prediction means (for example, the prediction means 12c shown in Figures 1 and 2) capable of predicting the sales floor situation of a certain store and the sales situation of a certain product in a certain store based on the relationship between the sales floor image data and the purchase history information in a certain store.

[0083] Furthermore, the AI ​​agent according to this embodiment (for example, the AI ​​agent 12 shown in Figure 1, the system terminal 12 shown in Figure 2) is an AI agent that acts as a substitute for a human and has the ability to learn and make decisions on its own, and the AI ​​agent, which is a computer, is configured to function as an image acquisition means (for example, the image acquisition means 12a shown in Figures 1 and 2) capable of acquiring store floor image data that captures the sales floor situation of a store, a learning means (for example, the learning means 12b shown in Figures 1 and 2) that learns the relationship between at least the store floor image data and purchase history information which is information about the history of products purchased by customers, and a prediction means (for example, the prediction means 12c shown in Figures 1 and 2) capable of predicting the sales floor situation of a certain store and the sales situation of a certain product in a certain store from one of the other based on the relationship between the store floor image data and the purchase history information in a certain store, and the learning means is configured to learn the relationship between the store floor image data and the purchase history information by being repeatedly given the store floor image data and the purchase history information as learning data.

[0084] According to the information processing system, information processing method, information processing program, and AI agent of this embodiment, it is possible to predict the sales status of products and the sales floor conditions of stores such as drugstores, and to replenish products displayed on the sales floor and implement sales strategies for products at the appropriate time.

[0085] Furthermore, the system may include means for acquiring external factors related to a particular store (for example, weather information for the area including the location of the store, the presence or absence of competing stores, the selling price of goods at competing stores, information on events or school activities in the area where the store is located, disaster information in the area where the store is located, and information on the trade area and population of the area where the store is located), and the prediction means may predict the sales status of the goods at the store based on the sales floor conditions of the store and the external factors of the store.

[0086] With this configuration, it becomes possible to predict the sales performance of a particular store by taking into account external factors related to that store, thereby improving the accuracy of the sales performance prediction for that store.

[0087] <<Embodiment 2>> The information processing system 10 according to Embodiment 2 of the present invention will be described below with reference to Figures 4 to 6.

[0088] <Overview of the Information Processing System> First, an overview of the information processing system 10 according to this second embodiment will be described using Figure 4. Figure 4 is a schematic diagram showing an overview of the information processing system 10 according to this second embodiment.

[0089] The information processing system 10 according to this embodiment is an information processing system comprising an AI agent 12 and an imaging means 17 (e.g., a camera) that captures images of the store's sales floor that can recognize the sales floor conditions, wherein the AI ​​agent 12 comprises at least a generation means 12a that generates prediction data by associating image data of the sales floor acquired from the imaging means 17 with weather information of the area including the store's location on the same time axis, and a prediction means 12b that takes a weather forecast of the area including the location of a certain store as input and uses the prediction data to predict the sales floor conditions of the store.

[0090] According to the information processing system of this embodiment, it is possible to predict the availability of sales floor space in stores such as drugstores and replenish the products displayed on the sales floor at an appropriate time.

[0091] Here, "AI agent" refers to a system that acts as a substitute (agent) for a human, possessing the ability to learn and make decisions on its own, and integrating various AI technologies for solving complex problems.

[0092] AI agents can perform tasks based on external triggers (e.g., user prompts, sensor inputs) or internal triggers (e.g., activation by an internal scheduler, occurrence of anomalies).

[0093] For example, when a question (prompt) is given as an external trigger (event), the system executes tasks such as searching for an answer to the question or generating an answer, and outputs the answer to the question. Also, for example, when a problem (prompt) is given as an external trigger (event), the system executes tasks such as searching for a solution to the problem or generating a solution, and outputs the solution to the problem.

[0094] Furthermore, "prompt" refers to instructions or questions given by a user to a system, and includes, for example, AI prompts given to generating AI, and command prompts that give instructions through command statements.

[0095] The various processes performed by AI agents can also be performed by "agent-type AI."

[0096] Furthermore, the "trigger" is not limited to triggers provided by external events (e.g., user prompts, sensor inputs), but may also be triggers provided by internal events (e.g., activation by an internal scheduler, occurrence of an anomaly).

[0097] The various processes performed by AI agents can also be performed by "agent-type AI."

[0098] Here, "agent-based AI" refers to a system in which multiple AI agents work together. In agent-based AI, each AI agent takes on a specialized role to solve complex problems that a single AI agent cannot solve, exchanging information with each other to output answers to questions or solutions to problems.

[0099] A "sales floor image" refers to an image that allows the "sales floor conditions" of a store to be recognized. "Sales floor conditions" include the number, condition, or name of products displayed on shelves in the sales floor, the availability of shelves in the sales floor, the number, condition, or name of products placed near shelves in the sales floor, the layout of shelves in the sales floor, the number of shoppers in the sales floor, and the status of store staff in the sales floor.

[0100] Examples of "store images" include images that allow recognition of increases or decreases in the number of products displayed on shelves in the store, images that allow recognition of the number, condition, or product name of products displayed on shelves in the store, images that allow recognition of the availability of shelves in the store, images that allow recognition of the number, condition, or product name of products placed near shelves in the store, images that allow recognition of the layout of shelves in the store, images that allow recognition of the number of shoppers in the store, and images that allow recognition of the status of store staff in the store.

[0101] "Weather information" refers to information related to the weather. This includes current weather information and past weather information. Examples of "weather information" include weather conditions (sunny, cloudy, rainy, snowy, etc.), temperature (maximum temperature, minimum temperature), humidity, atmospheric pressure, precipitation, sunshine duration, wind direction, wind speed, etc.

[0102] "Predictive data" is data generated by associating image data of store locations with weather information for the area including the store's location, all on the same time axis. For example, this would be data generated by acquiring data that associates image data of store locations at a certain date and time with weather information for the area including the store's location at the same date and time, at certain intervals (e.g., hourly intervals) over a certain period (e.g., six months).

[0103] A "weather forecast" refers to information that predicts the weather conditions of a particular area. Examples of weather forecasts include forecasts of weather conditions (sunny, cloudy, rainy, snowy, etc.), temperature (maximum temperature, minimum temperature), humidity, atmospheric pressure, precipitation, sunshine duration, wind direction, and wind speed. Weather forecasts can also be classified into short-term forecasts (e.g., hourly, daily), medium-term forecasts (weekly), and long-term forecasts (monthly, seasonal) depending on the period over which the forecast is made.

[0104] Furthermore, the generation means 12a may generate forecasting data by associating not only image data of the sales floor and weather information, but also management information related to the operation of the store, on the same time axis.

[0105] With this configuration, adding management information to the forecasting data can improve the accuracy of predictions about store sales floor conditions, allowing for timely replenishment of products on display.

[0106] Here, "management information" refers to information related to the management of a seller that sells products. Examples of "management information" include: (1) sales and profit-related data (e.g., sales data, profit data), (2) sales and product management-related data (e.g., product master data, specifications, shelf layout data, sales performance and promotional effectiveness), (3) inventory and procurement-related data (e.g., inventory status, ordering and procurement data), (4) customer and marketing-related data (e.g., customer ID data (ID-POS), marketing campaign results), (5) store operation and personnel management-related data (e.g., store performance, staff shifts), (6) expense and financial-related data (e.g., store and headquarters expenses, financial indicators), (7) external environment data (e.g., market area / population data, competitor store location and pricing information, weather, temperature, and disaster information), (8) local event / school event information, and (9) instruction and communication-related data (e.g., work instructions and policies from superiors, reports and suggestions from the field, history of collaboration with headquarters).

[0107] Furthermore, the management information may include store sales information (for example, sales information obtained in real time from the store's POS terminal), the store images may include images that allow recognition of increases or decreases in a particular product displayed in the store (for example, images that allow recognition of the availability of display shelves in the store), and the weather information may include location weather information related to the weather at the store's location (for example, pinpoint weather information obtained in real time from a weather forecasting website).

[0108] This configuration allows for improved accuracy in predicting store sales conditions and enables the replenishment of products displayed on the shelves at the appropriate time.

[0109] Furthermore, when the prediction data is updated by the generation means 12a, the system may be configured to provide the updated prediction data as training data.

[0110] With this configuration, it becomes possible to predict the sales floor conditions of stores based on the latest forecasting data, further improving the accuracy of the store sales floor condition predictions.

[0111] <Example of system configuration> Next, an example of the configuration of the information processing system 10 according to this second embodiment will be described using Figure 5. Figure 5 is a system configuration diagram showing an example of the configuration of the information processing system 10 according to this second embodiment.

[0112] The information processing system 10 can be configured, for example, to include a system terminal 12 that controls the entire system, and a management information terminal 14, a web server 18, and an external terminal 16 that are mutually connected to the system terminal 12 via a network NW.

[0113] The system terminal 12 is a terminal that controls the entire information processing system 10, and is composed of conventionally known servers, personal computers, etc. In this example, the system terminal 12 is composed of one server, but it may be composed of multiple servers, personal computers, etc. The hardware configuration of the system terminal 12 and the programs that the system terminal 12 executes will be described later.

[0114] The business information terminal 14 is a terminal (storage means) for a retail store (e.g., a drugstore, retail store, convenience store, etc.) to store business information, and is composed of conventionally known servers, personal computers, etc.

[0115] In this example, the management information terminal 14 is configured with a single server, but it may also be configured with multiple servers or personal computers. Furthermore, the terminal (storage means) for storing management information may be an internal storage means (for example, the storage device 26 shown in Figure 5) connected to the system terminal 12 via a local network (for example, LAN or P2P).

[0116] Web server 18 is a server that provides a website that distributes weather information and weather forecasts. The type of web server 18 is not particularly limited, but examples include servers of public organizations (e.g., the Japan Meteorological Agency) or private companies that distribute weather information and weather forecasts.

[0117] External terminals 16 are terminals used by users of the information processing system 10 (for example, business owners of drugstores, retail stores, convenience stores, etc.), and consist of personal computers, tablets, smartphones, etc. The type of external terminal 16 is not particularly limited, but examples include smartphones, personal computers, tablets, etc. used by individuals.

[0118] The network NW is a line that allows system terminals 12, management information terminals 14, web servers 18, and external terminals 16 to communicate with each other, and is typically composed of a WAN (Wide Area Network), also known as the Internet. The network NW may be wired or wireless, a LAN (Local Area Network), a dedicated line such as a VPN (Virtual Private Network), or a combination of these lines.

[0119] <System Terminal / Hardware Configuration Example> Next, we will describe an example of the hardware configuration of system terminal 12.

[0120] As shown in Figure 5, the system terminal 12 is configured to include, for example, a CPU 21, and a ROM 22, RAM 23, external storage drive 25, storage device 26, input device 27, display device 28, communication unit 29, imaging means 17, etc., all connected to the CPU 21 via a bus.

[0121] The CPU 21 is a control means that controls the entire system terminal 12, and performs processes such as executing application programs and operating systems (OS) stored in ROM 22 and storage devices 26, and storing data and files necessary for program execution in RAM 23 and storage devices 26.

[0122] ROM22 is a storage means for storing basic I / O programs and various data, and is composed of, for example, PROM, flash memory, etc. RAM23 is a storage means for temporarily storing data, and is composed of, for example, SDRAM, DRAM, etc. External storage drive25 is a control means that can read and write data to recording media 24 such as magnetic tape, DVD, etc., and is composed of, for example, magnetic tape storage, DVD drive, etc.

[0123] The storage device 26 is a storage means for storing application programs, the OS, control programs, related programs, various information, etc., and is composed of, for example, a hard disk drive (HDD), a solid-state drive (SDD), etc. The input device 27 is for inputting commands (instructions), etc., to the system terminal 12, and is composed of, for example, a keyboard, a pointing device (mouse, etc.), a touch panel, etc.

[0124] The display device 28 displays commands input by the input device 27, the response output of the system terminal 12 to those commands, and various other displays, and is composed of, for example, a liquid crystal display, plasma display, organic EL, etc. The communication unit 29 is a control means that controls communication with the management information terminal 14, web server 18, external terminal 16, etc. via the network NW, and is composed of, for example, a communication card, etc.

[0125] The imaging means 17 is a means for capturing images of the store's sales floor that allow for recognition of the store's sales floor conditions, and in this example, it consists of a camera installed facing the display shelves in the store's sales floor.

[0126] <System Terminal / Function> Next, we will explain the functions of the system terminal 12.

[0127] The storage device 26 of the system terminal 12 stores a program (information processing program) that causes the system terminal 12 to function as a generation means 12a and a prediction means 12b.

[0128] <System terminal / Function / Generation method> Next, the generation means 12a will be described.

[0129] The generation means 12a is a means for generating forecast data by associating image data of the sales floor acquired from the imaging means 17 with weather information of the region including the store's location on the same time axis. In this example, it is composed of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0130] As mentioned above, a "sales floor image" refers to an image that allows the "sales floor situation" of a store to be recognized. The "sales floor situation" includes the number, condition, and name of products displayed on shelves in the sales floor, the availability of shelves in the sales floor, the number, condition, and name of products placed near shelves in the sales floor, the layout of shelves in the sales floor, the number of shoppers in the sales floor, and the status of store staff in the sales floor.

[0131] Examples of "store images" include images that can recognize the increase or decrease in a certain product on a display shelf in a store, images that can recognize the number, condition, or product name of a product on a display shelf in a store, images that can recognize the availability of a display shelf in a store, images that can recognize the number, condition, or product name of a product placed near a display shelf in a store, images that can recognize the layout of a display shelf in a store, images that can recognize the number of shoppers in the store, and images that can recognize the status of store staff in the store. Conventional image recognition techniques can be applied to recognize these store images.

[0132] "Weather information" refers to information related to the weather. This includes current weather information and past weather information. Examples of "weather information" include weather conditions (sunny, cloudy, rainy, snowy, etc.), temperature (maximum temperature, minimum temperature), humidity, atmospheric pressure, precipitation, sunshine duration, wind direction, wind speed, etc.

[0133] "Predictive data" is data generated by associating image data of store locations with weather information for the area including the store's location, all on the same time axis. For example, this would be data generated by acquiring data that associates image data of store locations at a certain date and time with weather information for the area including the store's location at the same date and time, at certain intervals (e.g., hourly intervals) over a certain period (e.g., six months).

[0134] The generation means 12a acquires image data of the sales floor from an imaging means 17 installed in a certain store, and also acquires weather information for the area including the location of the store from a web server 18. It generates forecast data by associating this sales floor image data and weather information with the date and time of acquisition, and stores it in the storage device 26 in association with the ID of the store.

[0135] The generation means 12a repeatedly performs the process of acquiring image data of a store's sales floor at a certain date and time, and weather information for the area including the store's location at the same date and time, and generating forecast data, at certain intervals (e.g., every hour) over a certain period (e.g., six months).

[0136] Furthermore, the generation means 12a generates forecasting data by associating image data of a store's sales floor and weather information with management information related to the store's operations obtained from the management information terminal 14, all on the same time axis.

[0137] As mentioned above, "management information" refers to information related to the management of a seller that sells products. Examples of "management information" include: (1) sales and profit-related data (e.g., sales data, profit data), (2) sales and product management-related data (e.g., product master data, specifications, shelf layout data, sales performance and promotional effectiveness), (3) inventory and procurement-related data (e.g., inventory status, ordering and procurement data), (4) customer and marketing-related data (e.g., customer ID data (ID-POS), marketing campaign results), (5) store operation and personnel management-related data (e.g., store performance, staff shifts), (6) expense and financial-related data (e.g., store and headquarters expenses, financial indicators), (7) external environment data (e.g., market area / population data, competitor store location and pricing information, weather, temperature, and disaster information), (8) local event / school event information, and (9) instruction and communication-related data (e.g., work instructions and policies from superiors, reports and suggestions from the field, and history of collaboration with headquarters).

[0138] In the process of generating prediction data, the generation means 12a learns the relationships between image data of a store's sales floor (unstructured data), weather information for the area including the store's location (structured data), and management information related to the store's operations (structured data). It determines that these data are related and acquires combinations of related structured and unstructured data (in this example, combinations of image data, weather information, and management information).

[0139] Here, "structured data" refers to data that follows a predetermined format or rules and consists of numbers, characters, symbols, or a combination thereof.

[0140] Examples of "structured data" include numerical data (e.g., sales data, inventory data), time information (e.g., date, time), personal information (e.g., name, address, phone number), product information (e.g., product code, price, category), location information (e.g., latitude, longitude), sensor measurements (e.g., temperature, humidity, pressure), business information of sellers who sell products (business data), and purchase information of buyers (customers) who purchase products (purchase data).

[0141] "Management information (management data)" includes "management numerical information (data related to management figures)," and "management numerical information" refers to information used to evaluate a company's performance through management indicators and financial indicators.

[0142] Examples of "management numerical information" include: (1) sales and profit-related data (e.g., sales data, profit data), (2) sales and product management-related data (e.g., product master data, specifications, shelf layout data, sales performance and promotional effectiveness), (3) inventory and purchasing-related data (e.g., inventory status, ordering and purchasing data), (4) customer and marketing-related data (e.g., customer ID data (ID-POS), marketing campaign results), (5) store operations and personnel management-related data (e.g., store performance, staff shifts), (6) expense and financial-related data (e.g., store and headquarters expenses, financial indicators), (7) external environment data (e.g., trade area / population data, competitor store location and pricing information, weather, temperature, and disaster information), (8) local event / school event information, and (9) instruction and communication-related data (e.g., work instructions and policies from superiors, reports and suggestions from the field, history of collaboration with headquarters).

[0143] Furthermore, among the "data related to management figures," (2) sales and product management-related data (e.g., product master data, specifications, shelf layout data, sales performance and promotional effects) may be referred to as "data related to sales promotion (sales promotion)," and among the "data related to management figures," (4) customer and marketing-related data (e.g., customer ID data (ID-POS), results of marketing measures) may be referred to as "data related to purchase history."

[0144] "Purchase information (purchase data)" includes "purchase history information (data related to purchase history)," and "purchase history information" refers to information about the history of products purchased by a specific individual (customer).

[0145] Examples of "purchase history information" include: (1) customer information (customer ID, gender, age, membership rank, etc.), (2) product information (product ID, product name, category, brand, unit price, etc.), (3) purchase information (purchase date, purchase time, purchase quantity, total amount, payment method, surveys, staff interaction records, etc.), (4) store information (store ID, store name, sales channel (store / EC), etc.), (5) campaign information (status of campaigns and events, coupon usage, point usage, discount rate, etc.), and (6) purchase frequency (number of purchases, average purchase interval, most recent purchase date, etc.).

[0146] "Unstructured data" refers to data that does not have a predetermined format or rules, and which consists of text, images, audio, video, or a combination thereof.

[0147] Examples of "unstructured data" include text (e.g., emails, messages, notes (e.g., notes written by product buyers), questionnaires), images (e.g., photographs, illustrations, electronic flyers (digital flyers)), audio (e.g., chats, conversations, music), and videos (videos captured by cameras, videos created with video editing apps). Figure 6 shows an example of prediction data for a certain store and a prediction of the sales floor situation for that store.

[0148] For example, the generation means 12a acquires, at a certain date and time T1, image data of a store's sales floor (for example, an image that can identify that the stock of beverages on the shelves in the sales floor is low), weather information for the area including the location of the store (for example, the weather is sunny, the maximum temperature is 30 degrees, the humidity is 60%, and the sunshine duration is 8 hours), and business information of the store (for example, that beverage sales are high).

[0149] Then, the image data of these sales floor images, weather information, and management information are associated with the date and time T1 upon acquisition to generate forecast data 1, which is then stored in the storage device 26 in association with the ID of a particular store.

[0150] Furthermore, the generation means 12a acquires, at a certain date and time T2 separate from a certain date and time T1, image data of a store's sales floor (for example, an image that can identify that there are enough products on the shelves of the sales floor), weather information for the area including the location of the store (for example, the weather condition is snow, the maximum temperature is 2 degrees, the humidity is 40%, and the sunshine duration is 0 hours), and business information of the store (for example, overall sales of products are low).

[0151] Then, the image data of these sales floor images, weather information, and management information are associated with the date and time T2 to generate forecast data 2, which is then stored in the storage device 26 in association with the ID of a particular store.

[0152] Furthermore, the generation means 12a acquires, at a certain date and time T3 separate from a certain date and time T2, image data of a store's sales floor (for example, an image that can identify that there is a low stock of hot food on the shelves of the sales floor), weather information for the area including the location of the store (for example, the weather is rainy, the maximum temperature is 10 degrees, the humidity is 90%, and the sunshine duration is 1 hour), and business information of the store (for example, that hot food sales are high).

[0153] Then, the image data of these sales floor images, weather information, and management information are associated with the date and time T3 to generate forecast data 3, which is then stored in the storage device 26 in association with the ID of a particular store.

[0154] <System terminal / Function / Prediction method> Next, the prediction means 12b will be described.

[0155] The prediction means 12b is a means that takes a weather forecast for a region including the location of a certain store as input and uses prediction data to predict the sales floor conditions of the store. In this example, it is composed of a program stored in the storage device 26 of the system terminal 12, and the storage device 26, etc.

[0156] As mentioned above, a "weather forecast" refers to information that predicts the weather conditions of a particular area. Examples of weather forecasts include forecasts of weather conditions (sunny, cloudy, rainy, snowy, etc.), temperature (maximum temperature, minimum temperature), humidity, atmospheric pressure, precipitation, sunshine duration, wind direction, and wind speed. Furthermore, weather forecasts can be classified into short-term forecasts (e.g., hourly, daily), medium-term forecasts (weekly), and long-term forecasts (monthly, seasonal) depending on the period over which the forecast is made.

[0157] The prediction means 12b obtains weather forecast information for a region including the location of a certain store from the web server 18, and also refers to the prediction data for a certain store stored in the storage device 26 by the generation means 12a to predict the sales floor conditions of the certain store, and stores the predicted sales floor conditions information in the storage device 26 in association with the ID of the certain store.

[0158] For example, consider a case where the weather forecast for the next day in the area including a certain store's location is "sunny weather, maximum temperature 32 degrees Celsius," and we use forecast data, including forecast data 1-3 shown in Figure 6, to predict the sales floor conditions of that store the next day.

[0159] In this case, the prediction means 12b predicts the sales floor situation of a certain store the following day based on the fact that the weather forecast for a certain store the next day (sunny weather, maximum temperature of 32 degrees Celsius) is similar to the weather information of the prediction data 1 (sunny weather, maximum temperature of 30 degrees Celsius, humidity of 60%, sunshine duration of 8 hours). It uses the image data of the sales floor of a certain store from the prediction data 1 (an image that can identify that there is a low stock of beverages on the shelves of the sales floor) and the management information of a certain store from the prediction data 1 (high beverage sales), and predicts, for example, that "the stock of beverages will be low."

[0160] This allows users of the information processing system 10 (for example, managers of drugstores, retail stores, convenience stores, etc.) to replenish the products displayed on the sales floor at the appropriate time by taking measures such as ordering more beverages or taking beverages out of the warehouse in advance, based on predictions of sales floor conditions such as "beverage stock running low."

[0161] Furthermore, consider a case where, for example, the weather forecast for the area including the location of a certain store is "snowy" one week in advance, and we use forecast data, including forecast data 1-3 shown in Figure 6, to predict the sales floor conditions of that store one week in advance.

[0162] In this case, the prediction means 12b predicts the sales floor situation of a certain store one week from now, based on the fact that the weather forecast for a certain store one week from now (weather condition: snow) is similar to the weather information of the prediction data 2 (weather condition: snow, maximum temperature: 2 degrees, humidity: 40%, sunshine duration: 0 hours). This prediction means 12b uses the image data of the sales floor image of a certain store from the prediction data 2 (an image that can be identified as showing that there are enough products on the shelves of the sales floor) and the management information of a certain store from the prediction data 2 (overall sales of products are low), and predicts, for example, that "there is sufficient product inventory."

[0163] As a result, users of the information processing system 10 (for example, managers of drugstores, retail stores, convenience stores, etc.) can maintain an appropriate number of products on display in the store by taking measures such as refraining from ordering products or reducing the number of products sent out from the warehouse, based on a prediction of the store situation that "there is sufficient stock of products."

[0164] Furthermore, consider a case where, for example, the average weather forecast for the rainy season in the area including the location of a certain store is "humidity of 80% and sunshine duration of 1 hour," and we use prediction data, including prediction data 1-3 shown in Figure 6, to predict the average sales floor conditions of a certain store during the rainy season.

[0165] In this case, the prediction means 12b predicts the average store conditions during the rainy season based on the fact that the average weather forecast for a certain store during the rainy season (humidity of 90%, sunshine duration of 1 hour) is similar to the weather information of the prediction data 3 (weather condition: rain, maximum temperature of 10 degrees, humidity of 90%, sunshine duration of 1 hour). It then uses the image data of the store's sales floor in the prediction data 3 (an image that can identify that there is a low stock of hot food on the shelves of the sales floor) and the management information of the store in the prediction data 3 (sales of hot food are low) to predict the average sales floor conditions for a certain store during the rainy season, and predicts, for example, that "the stock of hot food will decrease."

[0166] This allows users of the information processing system 10 (for example, owners of drugstores, retail stores, convenience stores, etc.) to replenish the products displayed on the sales floor at the appropriate time by taking measures such as ordering more hot food or taking out hot food from the warehouse in advance, based on predictions of sales floor conditions such as "the stock of hot food is running low."

[0167] According to this example, it is possible to predict the availability of shelves in stores such as drugstores and replenish the products displayed on the shelves at the appropriate time.

[0168] Furthermore, when the prediction data is updated by the generation means 12a, the system may be configured to provide the updated prediction data as training data.

[0169] With this configuration, it becomes possible to predict the sales floor conditions of stores based on the latest forecasting data, further improving the accuracy of the store sales floor condition predictions.

[0170] <Information Processing Systems / Summary> As described above, the information processing system according to this embodiment (for example, the information processing system 10 shown in Figures 4 and 5) is an information processing system configured to include an AI agent (for example, the AI ​​agent shown in Figure 4, the system terminal 12 shown in Figure 5) and an imaging means (for example, the imaging means 17 shown in Figures 4 and 5) for capturing images of the store's sales floor that can recognize the sales floor conditions of the store, wherein the AI ​​agent comprises at least a generation means (for example, the generation means 12a shown in Figures 4 and 5) for generating prediction data (for example, prediction data 1 to 3 shown in Figure 6) by associating the image data of the sales floor acquired from the imaging means with weather information of the region including the location of the store on the same time axis, and a prediction means (for example, the prediction means 12b shown in Figures 4 and 5) for taking a weather forecast of the region including the location of a certain store as input and using the prediction data to predict the sales floor conditions of the store (for example, the sales floor conditions of the store shown in Figure 6).

[0171] Furthermore, the information processing method according to this embodiment (for example, the method executed by the information processing system 10 shown in Figures 4 and 5) is an information processing method executed using an AI agent (for example, the AI ​​agent shown in Figure 4, the system terminal 12 shown in Figure 5) and an imaging means (for example, the imaging means 17 shown in Figures 4 and 5) that captures images of the store's sales floor that can recognize the sales floor conditions of the store, wherein the AI ​​agent comprises at least a generation step (for example, the generation means 12a shown in Figures 4 and 5) that generates prediction data (for example, prediction data 1 to 3 shown in Figure 6) by associating the image data of the sales floor acquired from the imaging means and weather information of the region including the location of the store on the same time axis, and a prediction step (for example, the prediction means 12b shown in Figures 4 and 5) that takes a weather forecast of the region including the location of a certain store as input and uses the prediction data to predict the sales floor conditions of the store (for example, the sales floor conditions of the store shown in Figure 6).

[0172] Furthermore, the information processing program according to this embodiment (for example, the program executed by the information processing system 10 shown in Figures 4 and 5) is a program for an information processing system configured to include an AI agent (for example, the AI ​​agent shown in Figure 4, the system terminal 12 shown in Figure 5) and an imaging means (for example, the imaging means 17 shown in Figures 4 and 5) that captures images of the store's sales floor situation, wherein the AI ​​agent (for example, the AI ​​agent 12 shown in Figure 4, the system terminal 12 shown in Figure 5), which is a computer, is configured to at least... This information processing program is characterized by comprising: a generation means (for example, generation means 12a shown in Figures 4 and 5) that generates prediction data (for example, prediction data 1 to 3 shown in Figure 6) by associating image data of the sales floor acquired from the imaging means with weather information of the region including the location of the store on the same time axis; and a prediction means (for example, prediction means 12b shown in Figures 4 and 5) that takes a weather forecast of the region including the location of a certain store as input and uses the prediction data to predict the sales floor conditions of that store (for example, the sales floor conditions of the store shown in Figure 6).

[0173] Furthermore, the AI ​​agent according to this embodiment (for example, the AI ​​agent 12 shown in Figure 4, the system terminal 12 shown in Figure 5) is an AI agent that acts as a substitute for a human and has the ability to learn and make decisions on its own. The AI ​​agent, which is a computer (for example, the AI ​​agent 12 shown in Figure 4, the system terminal 12 shown in Figure 5), is characterized in that it functions as a generation means (for example, the generation means 12a shown in Figures 4 and 5) that generates prediction data (for example, prediction data 1 to 3 shown in Figure 6) by associating image data of a store's sales floor that can recognize the sales floor situation of the store, acquired from an imaging means (for example, the imaging means 17 shown in Figures 4 and 5), and weather information of the region including the location of the store, on the same time axis, and as a prediction means (for example, the prediction means 12b shown in Figures 4 and 5) that takes a weather forecast for the region including the location of a certain store as input and uses the prediction data to predict the sales floor situation of the store (for example, the sales floor situation of the store shown in Figure 6), and when the prediction data is updated by the generation means, the updated prediction data is provided as training data.

[0174] According to the information processing system, information processing method, information processing program, and AI agent of this embodiment, it is possible to predict the availability of sales floor space in stores such as drugstores and replenish the products displayed on the sales floor at an appropriate time.

[0175] Furthermore, the generation means (or generation step) may generate the forecast data by associating the image data of the sales floor image and the weather information with management information related to the operation of the store on the same time axis.

[0176] This configuration allows for improved accuracy in predicting store sales conditions and enables the replenishment of products displayed on the shelves at the appropriate time.

[0177] Furthermore, the management information may include sales information of the store (for example, sales information obtained in real time from the store's POS terminal), the sales floor images may include images that allow recognition of increases or decreases in a certain product displayed in the store's sales floor (for example, images that allow recognition of the availability of display shelves in the sales floor), and the weather information may include location weather information relating to the weather at the store's location (for example, pinpoint weather information obtained in real time from a weather forecasting website).

[0178] With this configuration, adding management information to the forecasting data can improve the accuracy of predictions about store sales floor conditions, allowing for timely replenishment of products on display.

[0179] Furthermore, when the prediction data is updated by the generation means (or the generation step), the updated prediction data may be provided as training data.

[0180] With this configuration, it becomes possible to predict the sales floor conditions of stores based on the latest forecasting data, further improving the accuracy of the store sales floor condition predictions.

[0181] Furthermore, the data structure according to this embodiment is a data structure for data used by an AI agent (for example, the AI ​​agent shown in Figure 4, the system terminal 12 shown in Figure 5), and includes structured data which is data that conforms to a predetermined format or rules (for example, weather information and business information shown in Figure 6, numerical data (for example, sales data, inventory data), time information (for example, date, time), personal information (for example, name, address, telephone number), product information (for example, product code, price, category), location information (for example, latitude, longitude), sensor measurements (for example, temperature, humidity, pressure), business information of a seller that sells products (business data), and purchase information of a buyer (customer) that purchases products (purchase data)), and unstructured data which is data that does not conform to a predetermined format or rules (for example The data structure is characterized in that it includes image data as shown in Figure 6, text (e.g., emails, messages, memos (e.g., memos written by product buyers), questionnaires), images (e.g., photographs, illustrations, electronic flyers (digital flyers)), audio (e.g., chats, conversations, music), and videos (videos captured by a camera, videos created with a video creation application)), and the agent learns the relationships between the structured data and the unstructured data, acquires combinations of the structured data and the unstructured data that have relationships, and is used in processing to execute processing for an activation trigger (e.g., a prompt input by an external terminal 16) based on the acquired combinations of structured data and the unstructured data when an activation trigger is given.

[0182] According to the data structure of this embodiment, processing can be performed in response to triggers such as user prompts based on multiple types of data with different structures, thereby improving the quality and speed of processing by the agent.

[0183] Furthermore, the combination of structured data and unstructured data may be a combination of data that has a relationship based on the time axis.

[0184] With this configuration, it is possible to combine multiple types of data with different structures based on the time axis, allowing for effective utilization of the data.

[0185] Furthermore, the structured data may include weather data (for example, the weather information shown in Figure 6) which is weather information for a region including the location of a certain store, and management data (for example, the management information shown in Figure 6) which is management information related to the operation of the certain store, while the unstructured data may include image data of a store floor that allows recognition of the store's sales floor conditions (for example, the image data shown in Figure 6).

[0186] With this configuration, weather data, business data, and image data with different structures can be combined based on the time axis, allowing for effective utilization of the data.

[0187] Furthermore, the agent may be used in a process that generates forecast data (for example, forecast data 1 to 3 shown in Figure 6) by associating the image data, weather data, and business data on the same time axis, and takes the weather forecast for the area including the location of a certain store as input, and uses the forecast data to predict the sales floor conditions of the store (for example, the sales floor conditions of the store shown in Figure 6).

[0188] With this configuration, it becomes possible to predict the availability of shelves in stores such as drugstores and replenish the products displayed on the shelves at the appropriate time.

[0189] Furthermore, the agent may be an AI agent or an agent-type AI.

[0190] Furthermore, the agent may learn the relationship between the trigger and the combination of structured data and unstructured data through machine learning.

[0191] This configuration allows for further improvements in the quality and speed of processing performed by the agent.

[0192] It should be noted that the information processing system, information processing method, information processing program, and AI agent according to the present invention are not limited to the embodiments described above, and various modifications can be made without departing from the spirit of the present invention.

[0193] Therefore, for example, a part of the configuration of Embodiment 1 described using Figures 1 to 3 may be combined with a part of the configuration of Embodiment 2 described using Figures 4 to 6.

[0194] Furthermore, the prediction means 12c according to the above embodiment 1 may predict only the sales status of a product in a certain store based on the sales floor conditions of that store, or it may predict only the sales floor conditions of a certain store based on the sales status of a product in a certain store.

[0195] Furthermore, while Embodiment 2 above shows an example of generating forecasting data using management information related to store operations, forecasting data may also be generated using image data of the sales floor acquired from the imaging means and weather information for the area including the store's location, without using management information.

[0196] Furthermore, while Embodiment 2 above shows an example of generating forecasting data using sales information from management information, the present invention is not limited thereto. For example, forecasting data may be generated using inventory and purchasing-related data (e.g., inventory status, order and purchase data) from management information.

[0197] With this configuration, it is possible to predict store conditions by taking into account product inventory status and ordering / procurement data. For example, even if the data used for prediction is image data of the store (an image that can identify that there is low stock of beverages on the shelves), if there is management information (that there is sufficient stock of beverages), it is possible to predict the store condition as "sufficient stock of beverages," and measures such as refraining from ordering beverages can be taken.

[0198] Furthermore, the prediction means may be configured to generate an image that visualizes the sales floor situation of a certain store as predicted, or it may be configured to output the image to an external device (for example, an external terminal 16).

[0199] Furthermore, while the generation means 12a has shown an example of acquiring a combination of image data, weather information, and business information as a combination of related structured and unstructured data, the present invention is not limited thereto. For example, electronic flyers (digital flyers) and notes written by product buyers may be added as unstructured data related to these data, and five types of data may be acquired as a combination of related structured and unstructured data. [Industrial applicability]

[0200] The information processing system, information processing method, information processing program, and AI agent according to the present invention can be widely applied to fields such as retail, service, and manufacturing. [Explanation of Symbols]

[0201] 10 Information Processing Systems 12 System Terminals 12a Generation means 12b Prediction means 16 External terminals 17. Imaging means 18 Web Server 21 CPU 22 ROM 23 RAM 24 Recording media 25 External storage drives 26 Storage device 27 Input devices 28 Display device 29 Communications Department

Claims

1. An information processing system comprising an AI agent and a storage means capable of storing purchase history information, which is information about the history of products purchased by a customer, The aforementioned AI agent, An image acquisition means capable of acquiring sales floor image data that captures the sales floor conditions of a store, A learning means for learning the relationship between at least the sales floor image data and the purchase history information, The system includes a prediction means capable of predicting the sales floor conditions of a certain store and the sales performance of a certain product in a certain store based on the relationship between the sales floor image data and the purchase history information of a certain store, The aforementioned sales floor image data includes image data that allows for the recognition of the inventory status of products displayed on the sales floor of the store. The learning means at least learns the relationship between the inventory status information of a certain product obtained from the sales floor image data and the purchase quantity of the same product obtained from the purchase history information. The prediction means identifies the current inventory status of a certain product by referring to the sales floor image data, and predicts the sales status of the certain product using the purchase quantity of the certain product associated with the inventory status information. An information processing system characterized by the following:

2. In the information processing system described in Claim 1, The prediction means identifies the current purchase quantity of a certain product by referring to the purchase history information, and predicts the sales floor status of the certain product using information on the inventory status of the certain product associated with the purchase quantity of the certain product. An information processing system characterized by the following:

3. In the information processing system according to claim 1 or 2, The system includes means for acquiring external factors related to the store, The aforementioned external factors include weather information for the area where the store is located. The system includes a generation means for generating forecast data by associating the aforementioned sales floor image data of a certain store, weather information for the region including the location of the certain store, and the aforementioned purchase history information of the certain store on the same time axis, The prediction means acquires weather forecast information for a region including the location of a certain store, and when the acquired weather forecast information is similar to the weather information included in the prediction data for the certain store, it predicts the sales floor conditions of the certain store using the sales floor image data and purchase history information associated with the weather information. An information processing system characterized by the following:

4. An information processing method that uses a computer, which is an AI agent, and a storage means capable of storing purchase history information, which is information about the history of products purchased by a customer, The aforementioned AI agent, An image acquisition step that can obtain sales floor image data capturing the sales floor situation of a store, A learning step that learns at least the relationship between the sales floor image data and the purchase history information, A prediction step is performed that allows for the prediction of one of the sales floor conditions and the sales status of the products in a certain store, based on the relationship between the sales floor image data and the purchase history information in a certain store, The aforementioned sales floor image data includes image data that allows for the recognition of the inventory status of products displayed on the sales floor of the store. The learning step at least learns the relationship between the inventory status information of a certain product obtained from the sales floor image data and the purchase quantity of the same product obtained from the purchase history information. The prediction step involves identifying the current inventory status of a certain product by referring to the store image data, and predicting the sales status of the certain product using the purchase quantity of the certain product associated with the inventory status information. An information processing method characterized by the following:

5. In the information processing method described in claim 4, The prediction step involves identifying the current purchase quantity of a certain product by referring to the purchase history information, and predicting the sales floor status of the certain product using the inventory status information of the certain product associated with the purchase quantity of the certain product. An information processing method characterized by the following:

6. A program for an information processing system comprising an AI agent and a storage means capable of storing purchase history information, which is information about the history of products purchased by a customer, The aforementioned AI agent, which is a computer, An image acquisition means capable of acquiring sales floor image data that captures the sales floor conditions of a store, A learning means for learning the relationship between at least the sales floor image data and the purchase history information, Based on the relationship between the sales floor image data and the purchase history information at a certain store, it functions as a predictive means capable of predicting the sales floor situation at a certain store and the sales status of the products at a certain store from one of them. The aforementioned sales floor image data includes image data that allows for the recognition of the inventory status of products displayed on the sales floor of the store. The learning means at least learns the relationship between the inventory status information of a certain product obtained from the sales floor image data and the purchase quantity of the same product obtained from the purchase history information. The prediction means identifies the current inventory status of a certain product by referring to the sales floor image data, and predicts the sales status of the certain product using the purchase quantity of the certain product associated with the inventory status information. An information processing program characterized by the following features.

7. In the information processing program described in claim 6, The prediction means identifies the current purchase quantity of a certain product by referring to the purchase history information, and predicts the sales floor status of the certain product using information on the inventory status of the certain product associated with the purchase quantity of the certain product. An information processing program characterized by the following features.

8. An AI agent that acts as a substitute for a human, and has the ability to learn and make decisions on its own, The aforementioned AI agent, which is a computer, An image acquisition means capable of acquiring sales floor image data that captures the sales floor conditions of a store, A learning means for learning the relationship between at least the aforementioned store image data and purchase history information, which is information about the history of products purchased by the customer, Based on the relationship between the sales floor image data and the purchase history information at a certain store, it functions as a predictive means capable of predicting the sales floor situation at a certain store and the sales status of the products at a certain store from one of them. The aforementioned sales floor image data includes image data that allows for the recognition of the inventory status of products displayed on the sales floor of the store. The learning means at least learns the relationship between the inventory status information of a certain product obtained from the sales floor image data and the purchase quantity of the same product obtained from the purchase history information. The prediction means identifies the current inventory status of a certain product by referring to the sales floor image data, and predicts the sales status of a certain product using the purchase quantity of that product associated with the inventory status information of that product. The learning means is configured to learn the relationship between the sales floor image data and the purchase history information, which are repeatedly provided as learning data. An AI agent characterized by the following.

Images